Comment by aneth4 | Hacker News Reader

Comment by aneth4 | Hacker News Reader

Parent Full thread

> Every major idea?

Ok, maybe not even almost every major idea (I did use the word "almost"), but pretty much by definition every major conceptual shift, from a round earth, to relativity, to microorganisms.

> Scientific ideas tend to me testable. A lot of what this guys suggests is entirely unfalsifiable. He may as well be discussing Pastafarianism.

Scientific ideas do tend to be testable. This doesn't mean we can't have ideas that are not testable or not yet testable. Hopefully, every idea will some day be falsifiable, but we can't rely on that. We may need to speculate in untestable territory before we find a plausible and reasonable theory that is testable. And as you know, science does not tell us that anything is true, only that some things can't be proven false.

There is more to life than testable scientific understandings. I could love my mother or I could not love my mother. Can you test that? Because you can not test it, does it mean it should not be mentioned and is not important?

Human love, creativity, and consciousness are perhaps the most interesting and important parts of existence, yet they are not currently testable from an experiential viewpoint, and they may never be.

There are qualities of the consciousness, which by definition are qualities of existence - love, color, taste, beauty, suffering, pain, joy - which can only be explained scientifically in their mechanical reflections in brain matter. Science can not tell us anything about the qualitative conscious experience of these qualities are, yet those qualities may be the most important part of the universe. What would the universe be without the conscious experience of its wonder and beauty?

> There is more to life than testable scientific understandings. I could love my mother or I could not love my mother. Can you test that? Because you can not test it, does it mean it should not be mentioned and is not important?

No, but it does mean that it's not science.

>I could love my mother or I could not love my mother. Can you test that?

In theory, yes. What happens inside your brain isn't untestable or sacred, we're just not there yet technologically (but we'll get there).

Love is a subjective experience of consciousness. We must first have some clue about what consciousness is, then perhaps we will be able to scientifically explain what love is.

If I create a device that predicts with 99% accuracy whether a test subject loves their mother (by looking at the subject's brain), without ever answering any of your questions about consciousness, would you retract the above statement?

No, because you would not be explaining what love is, any more than explain what pain is by showing me the brain patterns the light up when I twist your nipple.

Whether of not you can test for love is immaterial to explaining the experience.

When did I say I want to be explaining what love is? I said it's testable in theory, and you don't seem to be arguing against that.

> When did I say I want to be explaining what love is?

Well, you were asking me to retract this statement based on the theoretical existence of a test:

> Love is a subjective experience of consciousness. We must first have some clue about what consciousness is, then perhaps we will be able to scientifically explain what love is.

Anyhow, I was using it as an example of something that we can't currently test for, not asserting that it will never be testable. So no, I'm not arguing that it will never be testable.

If we do test, however, it will most likely be a test for the brain activity associated with love, not for the actual qualitative feeling of love. As I've said, we know nearly nothing about what subjective experience is despite all we know being subject to it.

As far as science is concerned, there's nothing beyond being able to objectively model a phenomenon and make useful predictions about it. If I can fully model the set of brain processes that humans typically call "love", and if I can make useful and accurate predictions based on my model, that's all that I care about as a scientist.

Now, it's entirely possible that reasoning about "love" scientifically requires several layers of new language / models on top of the currently established models in neuroscience, which is why I said it's _theoretically_ possible.

Subjective experience exists objectively, as in, the subjective experience in your consciousness is an objective reality of the universe. We do not yet know how to model such things, and making useful predictions based in brain activity is great, but it will not be a remotely satisfying model of consciousness.

Reply on news.ycombinator.com